@pellux/goodvibes-agent 0.1.105 → 0.1.107

package/CHANGELOG.md

@@ -2,6 +2,24 @@
 
 All notable changes to GoodVibes Agent will be recorded here.
 
+## 0.1.107 - Unreleased
+
+- Add `goodvibes-agent launch` and `goodvibes-agent start` as explicit aliases for starting the Agent TUI.
+- Open the Agent operator workspace automatically on normal TUI launch after first-run setup has been shown.
+- Document the Bun global install path with `goodvibes-agent launch` as the first interactive smoke.
+
+## 0.1.106 - 2026-06-01
+
+- b67a2fe Polish Agent setup entrypoint
+- Harden Bun global install with `bun add -g --trust`, zero-untrusted lifecycle verification, and packed TUI launch smoke.
+- Make Agent status and doctor output client/operator focused, including external runtime diagnostics and fail-closed isolated Agent Knowledge readiness.
+- Route `/memory` to Agent-local durable memory and move conversation-pinned compaction notes to `/session-memory`.
+- Remove copied file-edit undo/redo behavior from Agent conversation commands.
+- Remove stale hidden-share shortcut guidance and keep diff/file operations delegation-oriented.
+- Make provider onboarding generic across provider routes, subscription sessions, and credential inventory while keeping OpenAI as an optional quick-start path.
+- Remove internal audit phrase leakage from source/package-facing guards and neutralize non-Agent knowledge wording.
+- Shorten release workflow validation by relying on the single branch-CI gate run instead of duplicating release checks.
+
 ## 0.1.105 - 2026-06-01
 
 - 7ea88c8 Remove development panel surface

package/README.md

@@ -12,8 +12,9 @@ The Agent product connects to an already-running GoodVibes runtime. It does not
 Install the public alpha package with Bun:
 
 ```sh
-bun add -g @pellux/goodvibes-agent
+bun add -g --trust @pellux/goodvibes-agent
 goodvibes-agent --help
+goodvibes-agent launch
 goodvibes-agent status
 goodvibes-agent profiles templates
 goodvibes-agent personas list
@@ -31,10 +32,19 @@ export PATH="$(bun pm bin -g):$PATH"
 goodvibes-agent --help
 ```
 
-If Bun reports untrusted lifecycle dependencies, trust only the package and dependencies required by this package:
+`goodvibes-agent` and `goodvibes-agent launch` both start the interactive Agent TUI. On a fresh Agent home, the TUI opens Agent setup first.
+
+`--trust` lets Bun run the package lifecycle scripts required by the shipped SDK and parser dependencies during global install. A healthy install should report no pending lifecycle scripts:
+
+```sh
+bun pm -g untrusted
+```
+
+If you previously installed without `--trust`, run the targeted trust command once and then re-check:
 
 ```sh
 bun pm trust -g @pellux/goodvibes-agent @pellux/goodvibes-sdk core-js tree-sitter-css tree-sitter-javascript tree-sitter-json tree-sitter-python tree-sitter-typescript
+bun pm -g untrusted
 ```
 
 ## Source Usage
@@ -55,7 +65,7 @@ bun run package:install-check
 bun run publish:check
 ```
 
-Inside the Agent TUI, use `/agent`, `/home`, or `/operator` to open the operator workspace. It is the Agent-first fullscreen workspace for setup, status, knowledge, local memory/skills, work-plan/approval review, automation observability, and explicit build delegation to GoodVibes TUI.
+After setup has been shown once, the Agent TUI opens directly into the operator workspace. You can also reopen it with `/agent`, `/home`, or `/operator`. It is the Agent-first fullscreen workspace for setup, status, knowledge, local memory/skills, work-plan/approval review, automation observability, and explicit build delegation to GoodVibes TUI.
 
 Inside the workspace, use `/agent-profile guide` to author custom profile starters without leaving the Agent TUI. The guided flow lists starters, exports starter JSON, imports edited local starters, and creates isolated Agent profiles from them.
 

package/docs/getting-started.md

@@ -13,8 +13,9 @@ Agent does not launch the runtime for you.
 ## Install From Package
 
 ```sh
-bun add -g @pellux/goodvibes-agent
+bun add -g --trust @pellux/goodvibes-agent
 goodvibes-agent --help
+goodvibes-agent launch
 goodvibes-agent status
 goodvibes-agent personas list
 goodvibes-agent skills list
@@ -28,10 +29,17 @@ export PATH="$(bun pm bin -g):$PATH"
 goodvibes-agent --help
 ```
 
-If Bun requires lifecycle trust:
+`--trust` lets Bun run the lifecycle scripts required by the shipped SDK and parser dependencies. A clean global install should leave no pending lifecycle scripts:
+
+```sh
+bun pm -g untrusted
+```
+
+If you previously installed without `--trust`, run the targeted trust command once and then re-check:
 
 ```sh
 bun pm trust -g @pellux/goodvibes-agent @pellux/goodvibes-sdk core-js tree-sitter-css tree-sitter-javascript tree-sitter-json tree-sitter-python tree-sitter-typescript
+bun pm -g untrusted
 ```
 
 ## Run From Source
@@ -45,7 +53,9 @@ bun run dev
 
 `bun run dev` starts the Agent TUI. The same entrypoint backs the installed `goodvibes-agent` command.
 
-Once the TUI opens, run `/agent`, `/home`, or `/operator` to open the Agent operator workspace. That fullscreen workspace is the current front door for setup/config, knowledge status, local memory and skills, read-only work/approval/automation views, and explicit GoodVibes TUI build delegation.
+`goodvibes-agent` and `goodvibes-agent launch` both start the interactive Agent TUI. On a fresh Agent home, the TUI opens Agent setup first.
+
+After setup has been shown once, the TUI opens directly into the Agent operator workspace. You can also reopen it with `/agent`, `/home`, or `/operator`. That fullscreen workspace is the current front door for setup/config, knowledge status, local memory and skills, read-only work/approval/automation views, and explicit GoodVibes TUI build delegation.
 
 Use `/agent-profile guide` inside that workspace to walk through starter-profile authoring. It lists built-in and local starters, exports a JSON starter for editing, imports the edited starter back into this Agent home, and creates isolated profiles from the result.
 

package/docs/release-and-publishing.md

@@ -14,11 +14,11 @@ GoodVibes Agent's current installable public alpha version is recorded in `packa
 End users install and run GoodVibes Agent with Bun:
 
 ```sh
-bun add -g @pellux/goodvibes-agent
+bun add -g --trust @pellux/goodvibes-agent
 goodvibes-agent --help
 ```
 
-Do not add non-Bun install instructions for this product. The package is hosted on the public package registry, but the supported install and smoke path is Bun.
+Do not add non-Bun install instructions for this product. The package is hosted on the public package registry, but the supported install and smoke path is Bun with `--trust`, followed by `bun pm -g untrusted` reporting zero pending lifecycle scripts.
 
 ## Required Gates
 
@@ -38,7 +38,7 @@ git diff --check
 
 The GitHub release workflow publishes to npm only when the repository variable `PUBLISH_NPM` is `true` and the repository secret `NPM_TOKEN` is configured. Without those repository settings, the workflow still validates and creates the GitHub release, but npm publish must be run from a local environment with an exported token.
 
-Branch CI is the only workflow that runs the full test suite. It runs `bun run test` once for the release SHA. The release workflow must not run tests again; it verifies that branch CI passed for the exact checked-out SHA and then continues with non-test release validation.
+Branch CI is the only workflow that runs the full test suite and release gates. It runs `bun run test` once for the release SHA, along with typecheck, architecture, performance, build, publish, and packed install checks. The release workflow must not run tests or duplicate those gates; it verifies that branch CI passed for the exact checked-out SHA and then performs only package packing, GitHub Release creation, and optional npm publish.
 
 Do not add targeted `bun test` passes or separate release-only test gates to CI, release, or aggregate scripts. Tests that matter for release must be included in the single full branch-CI suite.
 
@@ -47,6 +47,8 @@ Also run the package install smoke from a packed artifact. It must prove:
 - the installed command is on `PATH`
 - the bin target is `bin/goodvibes-agent.ts`
 - the Bun shebang survives pack/install
+- Bun global install uses `bun add -g --trust`
+- `bun pm -g untrusted` reports zero pending lifecycle scripts after install
 - `goodvibes-agent --help` works
 - `goodvibes-agent --version` reports the package version
 - the installed TUI launches in a PTY and does not exit immediately

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pellux/goodvibes-agent",
-  "version": "0.1.105",
+  "version": "0.1.107",
   "private": false,
   "description": "GoodVibes personal operator assistant TUI with a proactive Agent product brain, isolated Agent Knowledge, local profiles, routines, skills, personas, and explicit build delegation.",
   "type": "module",

package/src/cli/completion.ts

@@ -1,5 +1,7 @@
 const COMMANDS = [
   'tui',
+  'launch',
+  'start',
   'run',
   'exec',
   'onboarding',

package/src/cli/help.ts

@@ -29,11 +29,11 @@ export function renderGoodVibesHelp(binary = 'goodvibes-agent'): string {
     `       ${binary} [OPTIONS] <COMMAND> [ARGS]`,
     '',
     'Commands:',
-    '  tui [path]                 Start the interactive Agent terminal UI (default)',
+    '  tui|launch [path]          Start the interactive Agent terminal UI (default)',
     '  run|exec [prompt]          Run non-interactively with text/json/stream-json output',
-    '  status                     Print config, provider, auth, and onboarding posture',
+    '  status                     Print config, provider, auth, and setup posture',
     '  doctor                     Print status plus setup warnings',
-    '  onboarding [status]        Open Agent onboarding, or print onboarding status',
+    '  setup|onboarding [status]  Open Agent setup, or print setup status',
     '  models [provider]          List/use/pin selectable models and recent model history',
     '  providers                  List/inspect/use provider config/auth posture',
     '  profiles                   Manage isolated Agent profile homes',
@@ -81,10 +81,11 @@ export function renderGoodVibesHelp(binary = 'goodvibes-agent'): string {
     '',
     'Examples:',
     `  ${binary}`,
+    `  ${binary} launch`,
     `  ${binary} --no-alt-screen`,
     `  ${binary} --cd ~/work/project --model openai:gpt-5.2`,
-    `  ${binary} onboarding`,
-    `  ${binary} onboarding status`,
+    `  ${binary} setup`,
+    `  ${binary} setup status`,
     `  ${binary} status`,
     `  ${binary} --runtime-url http://127.0.0.1:3421 status`,
     `  ${binary} models current`,
@@ -119,23 +120,23 @@ type CommandHelp = {
 
 const COMMAND_HELP: Record<string, CommandHelp> = {
   tui: {
-    usage: ['tui [path]', '[prompt]'],
+    usage: ['tui [path]', 'launch [path]', '[prompt]'],
     summary: 'Start the interactive Agent terminal UI. A prompt starts Agent with that prompt seeded.',
-    examples: ['', 'tui ~/work/project', '"summarize current tasks"'],
+    examples: ['', 'launch', 'tui ~/work/project', '"summarize current tasks"'],
   },
   run: {
     usage: ['run [prompt] [--output text|json|stream-json]', 'exec [prompt]'],
     summary: 'Run a single non-interactive agent turn and write the result to stdout.',
     examples: ['run "summarize the current project"', 'run --output json "list risks"', 'exec --output stream-json "check runtime status"'],
   },
-  onboarding: {
-    usage: ['onboarding', 'setup', 'onboarding status'],
-    summary: 'Open the setup wizard, or inspect whether onboarding has already been shown for this user.',
-    examples: ['onboarding', 'onboarding status'],
+  setup: {
+    usage: ['setup', 'setup status', 'onboarding', 'onboarding status'],
+    summary: 'Open Agent setup, or inspect whether setup has already been shown for this user.',
+    examples: ['setup', 'setup status', 'onboarding status'],
   },
   status: {
     usage: ['status', 'status --json', '--runtime-url http://127.0.0.1:3421 status'],
-    summary: 'Print Agent config, provider, auth, runtime connection, and onboarding posture.',
+    summary: 'Print Agent config, provider, auth, runtime connection, and setup posture.',
     examples: ['status', 'status --json', '--runtime-url http://127.0.0.1:3421 status'],
   },
   doctor: {
@@ -321,7 +322,7 @@ const COMMAND_HELP: Record<string, CommandHelp> = {
   },
   tasks: {
     usage: ['tasks list', 'tasks show <taskId>'],
-    summary: 'Inspect in-process runtime tasks. Agent blocks copied task submission; use run for one-shot work or delegate for explicit build/fix/review handoff.',
+    summary: 'Inspect in-process runtime tasks. Agent blocks runtime-owned task submission; use run for one-shot work or delegate for explicit build/fix/review handoff.',
     examples: ['tasks list', 'tasks show task-123', 'run "check provider readiness"', 'delegate "fix the failing tests"'],
   },
   bundle: {
@@ -344,7 +345,7 @@ const COMMAND_HELP: Record<string, CommandHelp> = {
 const HELP_ALIASES: Record<string, string> = {
   app: 'tui',
   exec: 'run',
-  setup: 'onboarding',
+  onboarding: 'setup',
   provider: 'providers',
   model: 'models',
   profile: 'profiles',

package/src/cli/management-commands.ts

@@ -279,7 +279,7 @@ export async function handleTasks(runtime: CliCommandRuntime): Promise<string> {
   const [sub = 'list', ...rest] = runtime.cli.commandArgs;
   if (sub === 'submit') {
     return [
-      'GoodVibes Agent blocks CLI task submission from the copied task workflow.',
+      'GoodVibes Agent blocks CLI task submission from the runtime-owned task workflow.',
       '  policy: do normal assistant work in the main Agent conversation or use `goodvibes-agent run <prompt>` for an explicit one-shot run.',
       '  build/fix/review: use `goodvibes-agent delegate <task>` for explicit GoodVibes TUI handoff.',
       '  result: no local task was started.',

package/src/cli/package-verification.ts

@@ -82,8 +82,8 @@ const PACKAGE_FACING_FORBIDDEN_TEXT = [
   ['Cloud', 'flare'].join(''),
   ['Open', 'Claw'].join(''),
   ['Her', 'mes'].join(''),
-  ['capabilities', ' audit'].join(''),
-  ['capabilities', ' command'].join(''),
+  ['cap', 'abilities audit'].join(''),
+  ['cap', 'abilities command'].join(''),
   ['near', '-fork'].join(''),
   ['Optional ', 'Browser Access'].join(''),
   ['Optional ', 'Other-Device Access'].join(''),
@@ -113,11 +113,11 @@ const PACKAGE_FACING_REQUIRED_TEXT: readonly {
   readonly path: typeof PACKAGE_FACING_TEXT_PATHS[number];
   readonly required: readonly string[];
 }[] = [
-  { path: 'README.md', required: ['/api/goodvibes-agent/knowledge'] },
+  { path: 'README.md', required: ['/api/goodvibes-agent/knowledge', 'bun add -g --trust @pellux/goodvibes-agent', 'bun pm -g untrusted'] },
   { path: 'docs/README.md', required: ['/api/goodvibes-agent/knowledge'] },
-  { path: 'docs/getting-started.md', required: ['/api/goodvibes-agent/knowledge'] },
+  { path: 'docs/getting-started.md', required: ['/api/goodvibes-agent/knowledge', 'bun add -g --trust @pellux/goodvibes-agent', 'bun pm -g untrusted'] },
   { path: 'docs/runtime-connection.md', required: ['/api/goodvibes-agent/knowledge'] },
-  { path: 'docs/release-and-publishing.md', required: ['/api/goodvibes-agent/knowledge'] },
+  { path: 'docs/release-and-publishing.md', required: ['/api/goodvibes-agent/knowledge', 'bun add -g --trust @pellux/goodvibes-agent', 'bun pm -g untrusted'] },
 ];
 const NON_COMMAND_ROUTE_ROOTS = new Set(['api', 'status']);
 

package/src/cli/parser.ts

@@ -8,6 +8,8 @@ import type {
 const COMMAND_ALIASES: Readonly<Record<string, GoodVibesCliCommand>> = {
   tui: 'tui',
   app: 'tui',
+  launch: 'tui',
+  start: 'tui',
   run: 'run',
   exec: 'run',
   e: 'run',

package/src/cli/service-posture.ts

@@ -61,9 +61,9 @@ export interface CliServicePosture {
 }
 
 const ENDPOINTS: readonly { readonly id: RuntimeEndpointId; readonly label: string; readonly enabledKey: string }[] = [
-  { id: 'controlPlane', label: 'runtime API', enabledKey: 'controlPlane.enabled' },
-  { id: 'httpListener', label: 'incoming webhook listener', enabledKey: 'danger.httpListener' },
-  { id: 'web', label: 'browser companion', enabledKey: 'web.enabled' },
+  { id: 'controlPlane', label: 'runtime connection', enabledKey: 'controlPlane.enabled' },
+  { id: 'httpListener', label: 'inbound events endpoint', enabledKey: 'danger.httpListener' },
+  { id: 'web', label: 'browser companion route', enabledKey: 'web.enabled' },
 ];
 
 interface CliServicePostureOptions {
@@ -184,7 +184,7 @@ export async function buildCliServicePosture(
   const issues: string[] = [];
 
   if (serverBackedEnabled && !config.enabled) {
-    issues.push('External runtime endpoints are configured, but Agent runtime ownership is disabled by design.');
+    issues.push('External runtime connection settings are present, but Agent runtime lifecycle ownership is disabled by design.');
   }
   if (config.enabled && !config.autostart) {
     issues.push('External runtime host config has autostart off.');
@@ -229,16 +229,17 @@ function yesNo(value: boolean): string {
 export function formatCliServicePosture(posture: CliServicePosture, json = false): string {
   if (json) return JSON.stringify(posture, null, 2);
   return [
-    'GoodVibes external runtime diagnostics',
+    'GoodVibes Agent runtime connection diagnostics',
     '  ownership: managed outside goodvibes-agent',
-    `  host config enabled: ${yesNo(posture.config.enabled)}`,
-    `  autostart config: ${yesNo(posture.config.autostart)}`,
-    `  restartOnFailure config: ${yesNo(posture.config.restartOnFailure)}`,
-    `  runtime host flag: ${yesNo(posture.config.daemonEnabled)}`,
+    '  Agent owns lifecycle: no',
+    `  external host config present: ${yesNo(posture.config.enabled)}`,
+    `  external host autostart config: ${yesNo(posture.config.autostart)}`,
+    `  external host restart config: ${yesNo(posture.config.restartOnFailure)}`,
+    `  legacy host switch present: ${yesNo(posture.config.daemonEnabled)}`,
     `  log: ${posture.log.path ?? 'n/a'} (${posture.log.exists ? 'present' : 'missing'})`,
     ...(posture.log.readError ? [`  log read error: ${posture.log.readError}`] : []),
     '',
-    'Runtime endpoints:',
+    'Runtime connection checks:',
     ...posture.endpoints.map((endpoint) =>
       `  ${endpoint.label}: enabled=${yesNo(endpoint.enabled)} ${endpoint.binding.hostMode} ${endpoint.binding.host}:${endpoint.binding.port} posture=${endpoint.bindPosture.label}${endpoint.reachable === undefined ? '' : ` reachable=${yesNo(endpoint.reachable)}`}`,
     ),

package/src/cli/status.ts

@@ -113,9 +113,9 @@ export function buildCliDoctorFindings(options: CliStatusOptions): readonly CliD
   const marker = options.onboardingMarkers?.effective;
   const serverBackedEnabled = daemonEnabled || controlPlaneEnabled || listenerEnabled || webEnabled;
   const networkFacingSurfaces = [
-    ['runtime API', controlPlaneEnabled, controlPlaneBinding],
-    ['incoming webhook listener', listenerEnabled, httpListenerBinding],
-    ['browser companion', webEnabled, webBinding],
+    ['runtime connection', controlPlaneEnabled, controlPlaneBinding],
+    ['inbound events endpoint', listenerEnabled, httpListenerBinding],
+    ['browser companion route', webEnabled, webBinding],
   ].filter(([, enabled, binding]) => isNetworkFacing(enabled, binding as typeof controlPlaneBinding));
 
   const findings: CliDoctorFinding[] = [];
@@ -162,7 +162,7 @@ export function buildCliDoctorFindings(options: CliStatusOptions): readonly CliD
         severity: 'warning',
         summary: 'Isolated Agent Knowledge route is not ready.',
         cause: `${options.externalRuntime.agentKnowledge.route} returned ${options.externalRuntime.agentKnowledge.kind}${options.externalRuntime.agentKnowledge.statusCode === null ? '' : ` (${options.externalRuntime.agentKnowledge.statusCode})`}.`,
-        impact: 'Agent Knowledge ask/search will not use any fallback wiki or HomeGraph segment; it will fail closed until the Agent route is available.',
+        impact: 'Agent Knowledge ask/search will not use any fallback wiki or non-Agent knowledge segment; it will fail closed until the Agent route is available.',
         action: 'Update/restart the external runtime to the Agent-compatible SDK and verify goodvibes-agent compat.',
       });
     }
@@ -173,8 +173,8 @@ export function buildCliDoctorFindings(options: CliStatusOptions): readonly CliD
       id: 'runtime-ownership-external',
       area: 'runtime',
       severity: 'warning',
-      summary: 'External runtime endpoints are configured while Agent runtime ownership is disabled by design.',
-      cause: 'One or more runtime API, listener, or web settings are enabled while service.enabled is false.',
+      summary: 'External runtime connection settings are present while Agent lifecycle ownership is disabled by design.',
+      cause: 'One or more runtime connection, inbound events, or browser companion settings are enabled while service.enabled is false.',
       impact: 'The external GoodVibes runtime must own availability for those endpoints; Agent will not start or enable them.',
       action: 'Manage runtime availability from GoodVibes TUI or the owning host, then use Agent for read-only diagnostics.',
     });
@@ -224,10 +224,10 @@ export function buildCliDoctorFindings(options: CliStatusOptions): readonly CliD
       id: 'onboarding-incomplete',
       area: 'onboarding',
       severity: 'warning',
-      summary: 'Onboarding has not been shown for this user.',
-      cause: 'No global user onboarding check marker was found.',
+      summary: 'Agent setup has not been shown for this user.',
+      cause: 'No global user setup check marker was found.',
       impact: 'Important runtime, network, provider, auth, or permission choices may still be implicit defaults.',
-      action: 'Run /onboarding in GoodVibes Agent or goodvibes-agent onboarding status to review setup state.',
+      action: 'Run /setup in GoodVibes Agent or goodvibes-agent setup status to review setup state.',
     });
   }
 
@@ -284,10 +284,10 @@ export function buildCliDoctorFindings(options: CliStatusOptions): readonly CliD
       id: 'network-http-listener-enabled',
       area: 'network',
       severity: 'warning',
-      summary: 'The incoming webhook listener is reachable beyond loopback.',
-      cause: `Incoming webhook listener is enabled on ${httpListenerBinding.host}:${httpListenerBinding.port} with ${httpListenerBinding.hostMode} binding.`,
+      summary: 'The inbound events endpoint is reachable beyond loopback.',
+      cause: `Inbound events endpoint is enabled on ${httpListenerBinding.host}:${httpListenerBinding.port} with ${httpListenerBinding.hostMode} binding.`,
       impact: 'External tools and devices may be able to reach incoming event endpoints.',
-      action: 'Keep listener secrets/signature checks configured for every enabled webhook endpoint.',
+      action: 'Keep inbound-event secrets/signature checks configured for every enabled endpoint.',
     });
   }
 
@@ -379,7 +379,7 @@ export function renderCliStatus(options: CliStatusOptions): string {
       ? `  operatorTokens: ${options.auth.operatorTokenPresent ? 'present' : 'missing'} (${options.auth.operatorTokenPath})`
       : '  operatorTokens: unknown',
     '',
-    'External Runtime:',
+    'Runtime Connection:',
     ...(externalRuntime ? [
       `  baseUrl: ${externalRuntime.baseUrl}`,
       `  reachable: ${yesNo(externalRuntime.reachable)}${externalRuntime.statusCode === null ? '' : ` (HTTP ${externalRuntime.statusCode})`}`,
@@ -393,11 +393,9 @@ export function renderCliStatus(options: CliStatusOptions): string {
     ]),
     '',
     'Runtime Ownership:',
-    '  Agent hosting: external only',
-    `  Agent starts runtime: no`,
-    `  legacy host config present: ${yesNo(serviceEnabled)}`,
-    `  legacy host autostart: ${yesNo(serviceAutostart)}`,
-    `  legacy host restart policy: ${yesNo(restartOnFailure)}`,
+    '  Agent role: client/operator TUI only',
+    '  hosting lifecycle: external',
+    '  Agent starts runtime: no',
     ...(options.service ? [
       `  platform: ${options.service.managed.platform}`,
       `  installed: ${yesNo(options.service.managed.installed)}`,
@@ -405,12 +403,6 @@ export function renderCliStatus(options: CliStatusOptions): string {
       `  definition: ${options.service.managed.path}`,
       `  log: ${options.service.log.path ?? 'n/a'} (${options.service.log.exists ? 'present' : 'missing'})`,
     ] : []),
-    '',
-    'Runtime Endpoint Diagnostics:',
-    bindLine('runtimeApi', controlPlaneEnabled, controlPlaneBinding),
-    bindLine('incomingWebhook', listenerEnabled, httpListenerBinding),
-    bindLine('browserCompanion', webEnabled, webBinding),
-    '',
     'Onboarding:',
     `  checked: ${marker?.exists ? 'yes' : 'no'}`,
     `  scope: ${marker?.scope ?? 'none'}`,
@@ -418,6 +410,18 @@ export function renderCliStatus(options: CliStatusOptions): string {
   ];
 
   if (options.doctor) {
+    lines.push(
+      '',
+      'External Runtime Config Signals:',
+      `  host config present: ${yesNo(serviceEnabled)}`,
+      `  host autostart: ${yesNo(serviceAutostart)}`,
+      `  host restart policy: ${yesNo(restartOnFailure)}`,
+      '',
+      'Endpoint Diagnostics:',
+      bindLine('runtimeApi', controlPlaneEnabled, controlPlaneBinding),
+      bindLine('incomingWebhook', listenerEnabled, httpListenerBinding),
+      bindLine('browserCompanion', webEnabled, webBinding),
+    );
     lines.push('', 'Warnings:');
     if (findings.length === 0) {
       lines.push('  none');
@@ -439,7 +443,7 @@ export function renderCliStatus(options: CliStatusOptions): string {
 export function renderOnboardingCliStatus(options: CliStatusOptions): string {
   const marker = options.onboardingMarkers?.effective;
   return [
-    'GoodVibes Agent onboarding status',
+    'GoodVibes Agent setup status',
     `  checked: ${marker?.exists ? 'yes' : 'no'}`,
     `  scope: ${marker?.scope ?? 'none'}`,
     `  source: ${marker?.payload?.source ?? 'n/a'}`,

package/src/cli/tui-startup.ts

@@ -100,6 +100,7 @@ export function applyInitialTuiCliState(options: {
 }): void {
   const { cli, input, commandRegistry, commandContext, shellPaths, render } = options;
   const globalOnboardingMarker = readOnboardingCheckMarker(shellPaths, 'user');
+  const seededPrompt = cli.flags.prompt ?? (cli.rawCommand === undefined && cli.positionals.length > 0 ? cli.positionals.join(' ') : undefined);
   if (cli.command === 'onboarding') {
     input.openOnboardingWizard({ mode: 'edit', reset: true });
   } else if (cli.command === 'sessions' && cli.commandArgs[0] === 'resume') {
@@ -109,9 +110,10 @@ export function applyInitialTuiCliState(options: {
     }
   } else if (!globalOnboardingMarker.exists) {
     input.openOnboardingWizard({ mode: 'new', reset: true });
+  } else if (cli.command === 'tui' && seededPrompt === undefined) {
+    input.openAgentWorkspace(commandContext);
   }
 
-  const seededPrompt = cli.flags.prompt ?? (cli.rawCommand === undefined && cli.positionals.length > 0 ? cli.positionals.join(' ') : undefined);
   if (seededPrompt) {
     input.prompt = seededPrompt;
     input.cursorPos = seededPrompt.length;

package/src/input/agent-workspace-categories.ts

@@ -27,14 +27,14 @@ export const AGENT_WORKSPACE_CATEGORIES: readonly AgentWorkspaceCategory[] = [
     detail: 'Agent connects to an existing GoodVibes runtime and owns local assistant configuration only. Runtime lifecycle and connectivity posture remain external.',
     actions: [
       { id: 'config', label: 'Open settings', detail: 'Open the fullscreen Agent settings workspace.', command: '/config', kind: 'command', safety: 'safe' },
-      { id: 'onboarding', label: 'Run setup', detail: 'Review Agent settings in the fullscreen setup flow.', command: '/onboarding', kind: 'command', safety: 'safe' },
+      { id: 'setup', label: 'Run setup', detail: 'Review Agent settings in the fullscreen setup flow.', command: '/setup', kind: 'command', safety: 'safe' },
       { id: 'setup-provider-model', label: 'Provider and model', detail: 'Choose the provider/model route for normal assistant chat.', command: '/model', kind: 'command', safety: 'safe' },
       { id: 'setup-agent-knowledge', label: 'Agent Knowledge', detail: 'Inspect the isolated Agent Knowledge store before ingesting source-backed material.', command: '/knowledge status', kind: 'command', safety: 'read-only' },
       { id: 'setup-runtime-profiles', label: 'Agent profiles', detail: 'Browse starter templates for isolated Agent homes and operator identities.', command: '/agent-profile templates', kind: 'command', safety: 'read-only' },
       { id: 'setup-personas', label: 'Personas', detail: 'Create or select the active local Agent persona.', targetCategoryId: 'personas', kind: 'workspace', safety: 'safe' },
       { id: 'setup-skills', label: 'Skills', detail: 'Create, review, and enable reusable local Agent skills.', targetCategoryId: 'skills', kind: 'workspace', safety: 'safe' },
       { id: 'setup-routines', label: 'Routines', detail: 'Create, review, and enable local Agent routines before any explicit schedule promotion.', targetCategoryId: 'routines', kind: 'workspace', safety: 'safe' },
-      { id: 'setup-memory', label: 'Local memory', detail: 'Inspect local/session memory; secrets stay rejected or redacted.', command: '/memory', kind: 'command', safety: 'read-only' },
+      { id: 'setup-memory', label: 'Local memory', detail: 'Inspect Agent-local memory; secrets stay rejected or redacted.', command: '/memory', kind: 'command', safety: 'read-only' },
       { id: 'setup-channels', label: 'Channels', detail: 'Open companion pairing and channel readiness setup.', command: '/pair', kind: 'command', safety: 'safe' },
       { id: 'setup-voice-media', label: 'Voice and media', detail: 'Open TTS/media settings for voice and image-capable Agent flows.', command: '/config tts', kind: 'command', safety: 'safe' },
       { id: 'provider', label: 'Provider status', detail: 'Review provider/model posture.', command: '/provider', kind: 'command', safety: 'read-only' },
@@ -66,7 +66,7 @@ export const AGENT_WORKSPACE_CATEGORIES: readonly AgentWorkspaceCategory[] = [
       { id: 'knowledge-ingest-url', label: 'Ingest URL', detail: 'Ingest a URL into Agent Knowledge only. Requires an explicit --yes command with a real URL.', command: '/knowledge ingest-url <url> --yes', kind: 'command', safety: 'safe' },
       { id: 'knowledge-import-bookmarks', label: 'Import bookmarks', detail: 'Import a browser bookmark export into Agent Knowledge only. Requires an explicit --yes command with a real path.', command: '/knowledge import-bookmarks <path> --yes', kind: 'command', safety: 'safe' },
       { id: 'knowledge-review-queue', label: 'Review queue', detail: 'Inspect source, item, and issue review work before accepting, rejecting, or resolving anything.', command: '/knowledge queue', kind: 'command', safety: 'read-only' },
-      { id: 'knowledge-consolidation', label: 'Consolidation review', detail: 'Inspect consolidation candidates and reports before running Agent Knowledge mutations.', command: '/knowledge candidates', kind: 'command', safety: 'read-only' },
+      { id: 'knowledge-sources', label: 'Source library', detail: 'List source-backed records already ingested into the isolated Agent Knowledge segment.', command: '/knowledge list --kind sources', kind: 'command', safety: 'read-only' },
       { id: 'knowledge-ask', label: 'Ask Agent knowledge', detail: 'Close this workspace and run /knowledge ask <question> or ask normally in chat.', kind: 'guidance', safety: 'read-only' },
     ],
   },
@@ -81,8 +81,8 @@ export const AGENT_WORKSPACE_CATEGORIES: readonly AgentWorkspaceCategory[] = [
       { id: 'tts-provider', label: 'Choose TTS provider', detail: 'Open provider/model routing for spoken responses through the settings flow.', command: '/config tts.provider', kind: 'command', safety: 'safe' },
       { id: 'tts-speak', label: 'Speak a prompt', detail: 'Submit a normal assistant turn and play the reply through configured live TTS. Close this workspace and provide real prompt text.', command: '/tts <prompt>', kind: 'command', safety: 'safe' },
       { id: 'image-attach', label: 'Attach image input', detail: 'Attach an image to the next assistant turn. Close this workspace and provide a real path and prompt.', command: '/image <path> <prompt>', kind: 'command', safety: 'safe' },
-      { id: 'browser-tools', label: 'Browser tools', detail: 'Inspect browser/tool readiness without starting inbound endpoints or mutating runtime state.', command: '/mcp servers', kind: 'command', safety: 'read-only' },
-      { id: 'mcp-browser', label: 'Browser MCP tools', detail: 'Inspect MCP servers and tools, including browser/automation roles, without mutating server setup.', command: '/mcp servers', kind: 'command', safety: 'read-only' },
+      { id: 'browser-tools', label: 'Browser tools', detail: 'Inspect browser/tool server readiness without starting inbound endpoints or mutating runtime state.', command: '/mcp servers', kind: 'command', safety: 'read-only' },
+      { id: 'mcp-tools', label: 'MCP tool inventory', detail: 'List available MCP tools, including browser and automation roles, without changing server trust or setup.', command: '/mcp tools', kind: 'command', safety: 'read-only' },
     ],
   },
   {

package/src/input/commands/guidance-runtime.ts

@@ -13,13 +13,13 @@ export function registerGuidanceRuntimeCommands(registry: CommandRegistry): void
           ctx.openOnboardingWizard({ mode: 'edit' });
           return;
         }
-        ctx.print('Use /onboarding to open the setup wizard.');
+        ctx.print('Use /setup to open Agent setup.');
         return;
       }
       if (sub === 'print') {
         ctx.print([
           'Welcome To GoodVibes Agent',
-          '  /onboarding         - open the onboarding wizard with current settings preloaded',
+          '  /setup              - open Agent setup with current settings preloaded',
           '  /agent              - open the Agent operator workspace',
           '  /knowledge          - inspect isolated Agent Knowledge status, ask, and search',
           '  /memory             - manage local Agent memory',

package/src/input/commands/health-runtime.ts

@@ -326,7 +326,7 @@ export function registerHealthRuntimeCommands(registry: CommandRegistry): void {
         '  /health remote',
         '  /health maintenance',
         '  /health repair <domain>',
-        '  /onboarding',
+        '  /setup',
       ].join('\n'));
     },
   });

package/src/input/commands/onboarding-runtime.ts

@@ -3,12 +3,13 @@ import { openOnboardingWizard } from './runtime-services.ts';
 
 export function registerOnboardingRuntimeCommands(registry: CommandRegistry): void {
   registry.register({
-    name: 'onboarding',
-    description: 'Open the onboarding wizard with current settings preloaded for review and editing',
+    name: 'setup',
+    aliases: ['onboarding'],
+    description: 'Open Agent setup with current settings preloaded for review and editing',
     usage: '',
     handler(_args, ctx) {
       openOnboardingWizard(ctx, { mode: 'edit', reset: true });
-      ctx.print('Opening onboarding wizard.');
+      ctx.print('Opening Agent setup.');
     },
   });
 }